On Modeling and Stability Analysis of Grid-Connected Voltage Source Converters
Time: Fri 2021-06-04 10.00
Location: zoom link for online defense (English)
Subject area: Electrical Engineering
Doctoral student: Hongyang Zhang , Elkraftteknik, Power Electronics
Opponent: Professor Xiaoming Yuan, Huazhong University of Science and Technology, Wuhan, China
Supervisor: Stefan Östlund, Elkraftteknik
With increasing deployment of voltage source converters (VSCs) in the grid, instability phenomena due to converter--grid interactions have been observed. Modeling and stability analysis of grid-connected VSCs have become popular research topics. Moreover, the power system is losing its inertia with increased penetration of renewable energy resources. Large and fast frequency deviations tend to vary more frequently compared to the conventional grid. The control and modeling of VSCs in the low-inertia grid is thus a crucial topic. The main objective of this thesis is the small-signal modeling and stability analysis of grid-connected VSCs.
As a result of the asymmetrical synchronous reference ($ dq $) frame dynamics, the structure of the dynamical system is multi-input multi-output (MIMO). Conventionally, the small-signal stability analysis for such a system is conducted either by the modal-analysis approach or the impedance based method, which is based on MIMO. In this thesis, methods for modeling and analyzing a MIMO system using a single-input single-output (SISO) approach are developed. Compared to conventional methods, the SISO methods overcome the difficulties of defining the stability margins for individual closed loops. Moreover, the methods provide intuitive solutions to the design of controllers for individual closed loops.
The thesis also investigates the control and modeling of grid-connected VSCs connected to low-inertia grids. It is illustrated that, for improvement of network frequency dynamics, a VSC with proper energy storage is capable to emulate inertial response from a synchronous machine. Moreover, other frequency control strategies are shown to be beneficial, compared to the inertial emulation. In addition, the inaccuracies for the $ dq $-frame small-signal model, in the presence of phase jumps and frequency deviations, are investigated. It is concluded that the inaccuracy of the small-signal model always exists under phase jumps. In the presence of frequency deviation, the stability analysis shall be evaluated at the steady-state operating point of the frequency.